Electron imaging methods have progressed to the point where single heavy atoms may be visualized in certain instances. Applications to biopolymer structure have been hampered by limitations related to specimen damage in the imaging electron beam, inadequate contrast in the specimen, and to the anhydrous nature of specimens. Since it now seems unlikely that wet naked biopolymers will be imaged at high resolution, further progress in preparation of specimens so as to minimize the effects of beam damage and dehydration is the only route to better utilization of microscopy as a tool in elucidating biological structure. At the same time, it is possible and important to solve pressing current problems in biopolymer structure with present methods, in blood component proteins, muscle proteins and cancer related glycoproteins. The work proposed here is divided into several parts: the optimization of specimen preparation through improvements in fine grain metal replication, and dehydration of macromolecular preparations by lyophilization of biopolymers from solutions designed to minimize freezing artifacts; several specific problems in biological structure: mapping of structural features of macromolecules by means of group specific markers (e.g., lectins or IgG complexed with glycoprotein glycosidic side chains - particularly with regard to epiglycanin, a mammary tumor cell surface glycoprotein which has been related to metastatic potential and which is too large for ordinary chemical mapping; myosin light chain loci; fibrinogen plasmin fragment loci); investigation of the structure and conformation of spectrin and its polymeric aggregates.